The Knowledge Ledger: Who Knew What, and When
A knowledge ledger records, for every secret in your series, who learned it, in which scene, and how. It catches the two errors readers argue about in forums: premature knowledge, where a character acts on something they were never told, and impossible ignorance, where a character forgets a thing the story showed them learning. Here is how to keep one.
Check 09 of the 12-check Series Continuity Self-Audit →
The letter at dinner
She hides the letter from everyone in Book 2. In Book 3, her brother references its contents at dinner, and no scene ever showed him learning them. Readers notice this one in forums, with page numbers, because a secret is a promise about who is in the dark, and the plot of a series is largely the schedule on which that dark lifts. One character who knows too early unravels a mystery; one who knows too late makes a scene impossible.
Keeping the ledger
One row per secret: the fact; who knows it at the start; each character who learns it, with the book, chapter and scene, and how (told, overheard, deduced, read); and who must never learn it before a named scene. Add a row when you draft the secret, not when you revise, and check the ledger before writing any scene in which a character acts on information. For a mystery, the ledger and the clue trail are the same document; the mystery clue and plot-thread tracker is the long version.
Premature knowledge and impossible ignorance
Premature knowledge is the brother at dinner: a character uses a fact no scene gave them. Impossible ignorance is its mirror: a character who was in the room when the fact was spoken asks about it, chapters later, as news. The second is the quieter error and the commoner one in long series, because a draft written before a scene was moved still carries the old schedule.
Intentional mysteries
A character who knows more than they should is also how twists work. The difference is on the page: an intentional mystery leaves a trace the re-reader can find (a glance, a hesitation, a scene that reads differently the second time), and the reveal names how they knew. A knowledge ledger holds those too, in a column for “knows, but the reader must not yet know they know”.
In a Full read (the Author, Series and Studio plans, and the free trial at public launch), the Coherence Guardian tracks what each character knows within one book and can flag a character who knows something they were never told. In our test manuscript (read on October 13, 2026), a character who showed knowledge of a hiding place that only the protagonist had read about was flagged as a character-knowledge issue, with the timing noted; the read then accepted a later line as his source, so treat a flag as a prompt to check the scene, not a verdict. A character asking, as news, about a fact she was shown learning was flagged in the same test. Across books, a cross-book check compares what each book records about a character’s appearance and voice and about an event’s dates, and nothing else: who knew what in Book 2 is not carried into the read of Book 3. That part of the ledger stays yours.
Questions authors ask
A character acting on information the story never gave them: quoting the letter they never read, going straight to the hideout nobody told them about.
It is one kind of plot hole: a break in the story’s own logic rather than a contradicted fact. How to find plot holes in your novel covers the family.
No. Inside one book, in a Full read, it can flag a character who knows something they were never told. Between books nothing tracks who knew what; the ledger is the tool for that.
StoryHelm is an AI manuscript-intelligence platform for indie series authors. It reads and analyzes an entire fiction series, catches continuity errors before readers do, and never writes or rewrites the author’s prose. You write every word of your book yourself, in StoryHelm’s built-in editor or in files you import. Two disclosed features produce text that isn’t your manuscript: Ship Dock drafts pitch material about the book (a query letter, synopsis, logline or back-cover blurb) for you to review and edit, and Translation Studio produces machine-generated translated editions of a finished book.
Keep reading
AI Plot Hole Checker for Novels
What a Full read can flag inside a book, a clue used too soon, and what it can’t.
ReadThe Resurrection Error: When the Dead Reappear
The resurrection error: something a story destroyed returns with no explanation. Why a series breeds it, a ledger that stops it, and what a read...
ReadMystery Series Clue and Plot-Thread Tracker
Track clues and plot threads in a mystery series. StoryHelm reads your books and flags dropped setups, broken alibis, and timeline contradictions.
ReadThe 5 Continuity Errors That Sink a Book Series
Continuity errors readers catch in a book series: name and trait drift, timeline slips, unpaid setups, voice drift and world-rule breaks, with a...
ReadEvery secret has a schedule. Keep it.
In a Full read, StoryHelm checks each book against its own canon and shows you where a character knows too early. The ledger, and the decision, stay yours. It never writes your prose.
Get the free Self-Audit